GB2585754A - Underwater image enhancement method and enhancement device - Google Patents
Underwater image enhancement method and enhancement device Download PDFInfo
- Publication number
- GB2585754A GB2585754A GB2006954.8A GB202006954A GB2585754A GB 2585754 A GB2585754 A GB 2585754A GB 202006954 A GB202006954 A GB 202006954A GB 2585754 A GB2585754 A GB 2585754A
- Authority
- GB
- United Kingdom
- Prior art keywords
- image
- underwater
- original
- sharp
- transmittance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000002834 transmittance Methods 0.000 claims abstract description 42
- 238000012634 optical imaging Methods 0.000 claims abstract description 28
- 230000004927 fusion Effects 0.000 claims abstract description 26
- 238000003384 imaging method Methods 0.000 claims abstract description 12
- 238000005070 sampling Methods 0.000 claims abstract description 9
- 230000008859 change Effects 0.000 claims description 10
- 239000003086 colorant Substances 0.000 claims description 4
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 2
- 238000007499 fusion processing Methods 0.000 abstract 1
- 230000000007 visual effect Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000005286 illumination Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000010521 absorption reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000003595 mist Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/46—Colour picture communication systems
- H04N1/56—Processing of colour picture signals
- H04N1/60—Colour correction or control
- H04N1/6077—Colour balance, e.g. colour cast correction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/80—Camera processing pipelines; Components thereof
- H04N23/84—Camera processing pipelines; Components thereof for processing colour signals
- H04N23/88—Camera processing pipelines; Components thereof for processing colour signals for colour balance, e.g. white-balance circuits or colour temperature control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N9/00—Details of colour television systems
- H04N9/64—Circuits for processing colour signals
- H04N9/646—Circuits for processing colour signals for image enhancement, e.g. vertical detail restoration, cross-colour elimination, contour correction, chrominance trapping filters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Image Processing (AREA)
Abstract
An underwater image enhancement method comprises: acquiring 101 an original underwater image, and establishing an underwater optical imaging model according to the characteristics of underwater imaging; sharpening 102 the original underwater image using a dark channel prior defogging algorithm that performs linear programming on the transmittance component of the model, to perform contrast enhancement and obtain a first sharp image; correcting 103 the colour of the original underwater image by using a gray world algorithm to obtain a second sharp image; and fusing 104 the first sharp image and the second sharp image by using an image fusion algorithm based on wavelet transform to obtain a restored underwater image. The contrast enhancement process to obtain the first sharpened image may comprise down-sampling the original underwater image, calculating a dark channel and transmittance image by using the down sampled image, and reconstructing the dark channel and transmittance image by means of bilinear interpolation. The image fusion process may comprise the use of a Haar wavelet basis and an inverse wavelet transform.
Description
UNDERWATER IMAGE ENHANCEMENT METHOD AND
ENHANCEMENT DEVICE
Field of the Invention
The present invention relates to the technical field of image enhancement and restoration, in particular to an underwater image enhancement method and enhancement device.
Background of the Invention
Due to the absorption and scattering of light caused by suspended particles in water and the different attenuation of light of different wavelengths under water, underwater images usually have problems such as blurred details, low contrast, and color distortion. Therefore, the quality of underwater images needs to be improved. Because an underwater optical imaging model is similar to a foggy imaging model to a certain degree, the existing dark channel prior defogging method can be referenced to eliminate underwater backscatter blurs and restore sharp images. However, unlike a foggy image with insignificant color shift, the water medium has great differences in the absorption characteristics of light of different wavelengths. The existing dark channel prior defogging algorithm has a deviation in the calculation of transmittance and is not suitable for underwater regions. When the intensity of pixels is close to the underwater ambient light value, the defogged image will have local color spots and color shift effect.
Summary of the Invention
The technical problem to be solved by the present invention is to provide an underwater image enhancement method and enhancement device to solve the problem of color shift effect in the prior art when the dark channel prior defogging method is used to restore an underwater image.
In order to solve the above technical problem, an embodiment of the present invention provides an underwater image enhancement method, including: acquiring an original underwater image, and establishing an underwater optical imaging model according to the characteristics of underwater imaging; sharpening the original underwater image by using the characteristic that the underwater optical imaging model is similar to an atmospheric model, and using a dark channel prior defogging algorithm that performs linear programming on the transmittance, to obtain a first sharp image, wherein the sharpening includes contrast enhancement; correcting the color of the original underwater image by using a gray world 10 algorithm to obtain a second sharp image; and fusing the first sharp image and the second sharp image by using an image fusion algorithm based on wavelet transform to obtain a restored underwater image. Further, the underwater optical imaging model is expressed as: 1(x) = J (x) * 1(x) + A* (1-1(x)) where 1(x) represents the original underwater image, J(.v) represents the first sharp image, 1(x) represents the transmittance of scene light, and A represents underwater ambient light.
Further, the sharpening formula used by the dark channel prior defogging algorithm is: 1(x)-A J(x)= +A max(t(x), t0) where to represents a transmittance threshold.
Further, 1(x) is newly linearly programmed to t(x)change = 1(x) * 0.9 + 0.1 where t(x),,, represents the newly programmed transmittance.
Further, sharpening the original underwater image by using the characteristic that the underwater optical imaging model is similar to an atmospheric model, and using a dark channel prior defogging algorithm that performs linear programming on the transmittance, to obtain a first sharp image includes: down-sampling the original underwater image; calculating a dark channel and transmittance image by using the reduced image obtained by the down-sampling as input; and reconstructing the dark channel and transmittance image by means of bilinear 5 interpolation to obtain the first sharp image with original image size.
Further, correcting the color of the original underwater image by using a gray world algorithm to obtain a second sharp image includes: multiplying three channels R, G, and B of the original underwater image by their respective gains through color temperature correction to adjust the ratio of three 10 colors R, G, and B to obtain: C(1)= R* C(0)=0 * o), C(B)= B* where c(R) C(0) and c(B) respectively represent corresponding values after the 15 three channels R, G, and B of the original underwater image are multiplied by their respective gains;' 2, and 6)3 represent gains, and the values of 6)1, -, and 6)3 are solved through the theory of maximum image information entropy according to the change of the image into a gray image: max(H(R * a), + (1 * co, + B co,))-> fro, ta, co where 11(.) represents an image entropy; and adjusting, for each pixel C in the image, its R, G, and B components according to the obtained 0(R) , (7(1), and 0(B) : 0(k)= C(R)* C(G)= 20)* kg 0(13)= 0(13)* kb where 0(k), C(G), and C(S) respectively represent the values obtained after 0(R) 0(0), and C(B) are multiplied by their respective gain coefficients, and kr, , kg, and kk, respectively represent the gain coefficients of the R, G and B channels Further, fusing the first sharp image and the second sharp image by using an image fusion algorithm based on wavelet transform to obtain a restored underwater image includes: performing three-layer wavelet decomposition on the first sharp image and the second sharp image by using Haar wavelet basis to obtain low-frequency 5 components and high-frequency components of different scales; processing the low-frequency components by average operator fusion and multiplying by a gain coefficient to obtain low-band wavelet coefficients; processing the high-frequency components by maximum regional energy fusion to obtain high-band wavelet coefficients; and it) reconstructing an image by inverse wavelet transform according to the obtained low-band wavelet coefficients and high-band wavelet coefficients to complete the fusion of the images and obtain the restored underwater image.
An embodiment of the present invention also provides an underwater image enhancement device, including: an establishment module, configured to acquire an original underwater image, and establish an underwater optical imaging model according to the characteristics of underwater imaging; a processing module, configured to sharpen the original underwater image by using the characteristic that the underwater optical imaging model is similar to an atmospheric model, and using a dark channel prior defogging algorithm that perfafins linear programming on the transmittance, to obtain a first sharp image, wherein the sharpening includes contrast enhancement; a correction module, configured to correct the color of the original underwater image by using a gray world algorithm to obtain a second sharp image; and a fusion module, configured to fuse the first sharp image and the second sharp image by using an image fusion algorithm based on wavelet transfonn to obtain a restored underwater image.
The beneficial effects of the above technical solution of the present invention are as follows: In the above solution, an original underwater image is acquired, and an underwater optical imaging model is established according to the characteristics of underwater imaging; the original underwater image is sharpened by using the characteristic that the underwater optical imaging model is similar to an atmospheric model, and using a dark channel prior defogging algorithm that performs linear programming on the transmittance, to obtain a first sharp image; the color of the original underwater image is corrected by using a gray world algorithm to obtain a second sharp image; and for the first dark sharp image corrected by the dark channel prior defogging algorithm and the second bright sharp image corrected by the gray world algorithm, the first sharp image and the second sharp image are fused by using an image fusion algorithm based on wavelet transfoim to obtain a restored underwater image with better contrast, color and brightness. In this way, the underwater image enhancement method based on multi-algorithm comprehensive optimization improves the visual effect of the underwater image, saves useful information of the image and reduces noise from the two aspects of contrast enhancement and color correction, so as to effectively improve the detail definition and color fidelity of the underwater image with low quality and low illumination, restore the real underwater scene, and then improve the accuracy and efficiency of underwater image restoration.
Brief Description of the Drawings
Fig. 1 is a schematic flowchart of an underwater image enhancement method according to an embodiment of the present invention; Fig. 2 is a schematic diagram of image enhancement results according to an 25 embodiment of the present invention; Fig. 3 is a schematic structural diagram of an underwater image enhancement device according to an embodiment of the present invention.
Detailed Description of the Embodim ents
In order to make the technical problems to be solved, technical solutions and advantages of the present invention more clear, detailed descriptions will be given below in combination with the accompanying drawings and specific embodiments.
The present invention provides an underwater image enhancement method and enhancement device against the problem of color shift effect when the existing dark channel prior defogging method restores an underwater image.
Embodiment 1 As shown in Fig. 1, an underwater image enhancement method according to an 10 embodiment of the present invention includes steps S101 to S104.
S101, an original underwater image is acquired, and an underwater optical imaging model is established according to the characteristics of underwater imaging; S102, the original underwater image is sharpened by using the characteristic that the underwater optical imaging model is similar to an atmospheric model, and using a dark channel prior defogging algorithm that performs linear programming on the transmittance, to obtain a first sharp image, wherein the sharpening includes contrast enhancement; S103, the color of the original underwater image is corrected by using a gray world algorithm to obtain a second sharp image; and S I 04, the first sharp image and the second sharp image are fused by using an image fusion algorithm based on wavelet transform to obtain a restored underwater image According to the underwater image enhancement method described in the embodiment of the present invention, an original underwater image is acquired, and an underwater optical imaging model is established according to the characteristics of underwater imaging; the original underwater image is sharpened by using the characteristic that the underwater optical imaging model is similar to an atmospheric model, and using a dark channel prior defogging algorithm that performs linear programming on the transmittance, to obtain a first sharp image; the color of the original underwater image is corrected by using a gray world algorithm to obtain a second sharp image; and for the first dark sharp image corrected by the dark channel prior defogging algorithm and the second bright sharp image corrected by the gray world algorithm, the first sharp image and the second sharp image are fused by using an image fusion algorithm based on wavelet transform to obtain a restored underwater image with better contrast, color and brightness. In this way, the underwater image enhancement method based on multi-algorithm comprehensive optimization improves the visual effect of the underwater image, saves useful info:Filiation of the image and reduces noise from the two aspects of contrast enhancement and color correction, so as to effectively improve the detail definition and color fidelity of the underwater image with low quality and low illumination, restore the real underwater scene, and then improve the accuracy and efficiency of underwater image restoration.
In a specific embodiment of the aforementioned underwater image enhancement method, further, the underwater optical imaging model is expressed as: 1(x) = 1(x) * I(x) + A (I -1(x)) Where /(x) represents the original underwater image, .1(x) represents the first sharp image, /(x) represents the transmittance of scene light, and A represents underwater ambient light.
In this embodiment, the existing dark channel prior model is mostly used to sharpen 20 foggy images. This prior model is based on an empirical assumption: in most sharp and fogless atmospheric images, the intensity value of at least one color channel is very low, even near 0, that is: 1(x)= min (nun (r (y)) -> 0 cc{r,g,b} yci-Mr) Where n(x) represents an x-centric image neighborhood, JE(,(x) is a dark channel of a fogless image I, C represents a color channel of the image, I' (y) represents a color channel of a point y of an RGB image, and in combination with the dark channel prior theory, the value of J:1k(x) is near 0 for a non-sky region of J Due to certain similarities between underwater images and foggy images, the prior model has also been gradually applied by scholars to underwater image restoration. In this embodiment, an original underwater image is acquired, and an underwater optical imaging model is established according to the characteristics of underwater imaging: 1 (x) = (x) * t (x) + A -t (x)) Where 1(x) represents the original underwater image, J(x) represents the first sharp image, ((x) represents the transmittance of scene light, and A represents underwater ambient light. In the underwater optical imaging model, At). i(t) represents an immediate component, and A (1-t(x)) represents a background 1() scattering component.
In this embodiment, when the transmittance t(x) and the underwater ambient light value A are known, they can be used to calculate a defogged image. However, considering that when the transmittance is near zero, the value of J(x) will tend to infinity, so that the overall restored fogless image will turn to a white field, resulting in image distortion, so a transmittance threshold to (generally 0.1) is required to limit the transmittance /(x) . The sharpening fonnula adopted by the dark channel prior defogging algorithm is: 1(x)-= +.4 max(t(x), t") In this embodiment, when j(r) is very close to A, t(v) will be very small, even close to 0, that is, t(x)co, and then, when Ale) is calculated, there will be many pixels corresponding to the same t(x) value, i.e., to, which will cause color spots. Because some problems occur when 1(r) is calculated, some processing can be done for 4') This embodiment proposes to newly linearly program the original transmittance 4') to t(x)th ge, where t(x)ge represents the newly programmed (r(y))) 1(x) =1-(-0 min( mill pec2(' transmittance, so that t(x) is always not less than 0.1.
where "x) represents an x-centric image neighborhood, c represents a color channel of the image, I (Y) represents a color channel of a point y of the original underwater image, and 4') represents an estimated transmittance value in a region L24) . In real life, even in fine weather, there are some impurities and particles in air, so when we look at distant scenes, we still feel a layer of mist If the mist is completely ignored, the resulting image will be unnatural and will lose the feeling of depth of field. Therefore, an impact factor o) between [0,1] is introduced to retain a certain degree of fog, generally m = 0.95, where Ac represents underwater ambient light of a color channel. The order of 0) values solved is unchanged, which avoids the situation that many pixels less than 0 1 in the original algorithm correspond to the same t(x) value. Thus, the value of original transmittance t(r) 10 is programmed from [0,1] to [0.1,1], and the calculated t(r) is linearly improved. In this way, by correcting the transmittance, the restored first sharp image will not have too high contrast, and the color shift effect can be reduced, that is, = 1(x)*0.9 +0.1 In a specific embodiment of the aforementioned underwater image enhancement method, further, the step that the original underwater image is sharpened by using the characteristic that the underwater optical imaging model is similar to an atmospheric model, and using a dark channel prior defogging algorithm, to obtain a first sharp image includes that: The original underwater image is down-sampled; A dark channel and transmittance image is calculated by using the reduced image obtained by the down-sampling as input; and The dark channel and transmittance image is reconstructed by means of bilinear interpolation to obtain the first sharp image with original image size.
The classic method used by the existing dark channel prior defogging algorithm in 25 the correction of a transmission image is a soft matting method. However, the soft matting method is time-consuming, and it takes a long time to process an image, which is a too long time for video processing In this embodiment, for the problem of large calculation amount and long processing time of the restored image, the dark channel and transmittance image calculated during defogging does not need high resolution, so the input original underwater image can be down-sampled first, and then A dark channel and transmittance image is calculated by using the reduced image obtained by the down-sampling as input; and finally, the dark channel and transmittance image is 5 reconstructed by means of bilinear interpolation to obtain the first sharp image with original image size. In this way, the calculation time can be effectively reduced, the time for subsequent processing can be saved, and the efficiency can be improved. In this embodiment, in order to better understand the present invention, the down-sampling and the bilinear interpolation will be described: it) 1) Down-sampling method An image I is assumed to be with a size of M*N and down-sampled by s times to obtain an image having a resolution of (41 I 8)*(N 6), where s is a common divisor of M and N. If it is a matrix image, the image within the original image S * 8 window (win) is changed into a pixel Pn, and the value of is a mean of all pixels within the window 01 71) . where / represents a certain point in the image I, and n represents a certain point within the window 2) Bilinear interpolation method For a target pixel, floating point coordinates obtained by inverse transformation of coordinates are set to be.1' (i + u, I + v) , where i and j are non-negative integers, u and v are numbers of floating points in the range [0, 11, and i, j, u, and v all represent image coordinate calculation values; then the value le + H,./ v) of this pixel can be determined by the values of four surrounding pixels corresponding to the coordinates (I, j) , (i, + I), (1+1,1), and (1+1, j +1) in the original image, that is: + + = -00 + 0 - j + 1) + 11(1 -v)f (i +1, j)+ int (i +1, j +1) In a specific embodiment of the aforementioned underwater image enhancement method, further, the step that the color of the original underwater image is corrected by using a gray world algorithm to obtain a second sharp image includes that: Three channels R, G, and B of the original underwater image are multiplied by their respective gains through color temperature correction to adjust the ratio of three colors R, G, and B to obtain: C(R) = R * C(G)=111*(0., C(B)= B* o3 Where "R) , and "B) respectively represent corresponding values after the three channels R, G, and B of the original underwater image are multiplied by their respective gains; 6-'^ , 6-2, and "', represent gains, and the values of "'^ , "2, and CD are solved through the theory of maximum image information entropy according to the change of the image into a gray image: max(H(R*()), + G * co + B * 3)) -> ( , t o,) where HO represents an image entropy; According to the obtained (7(R), (7(G), and (7(B), for each pixel C in the image, its 15 R, G, and B components are adjusted: C(R')= C(R) C(G*)= C(G)* kg )-(7(8) *k Where (AR.), (7(G), and c(B) respectively represent the values obtained after (7(R), C(G), and c(B) are multiplied by their respective gain coefficients, and A;, kg, and kb respectively represent the gain coefficients of the R, G and B channels. In this embodiment, the first sharp image, obtained by sharpening the original underwater image using the dark channel prior defogging algorithm that performs linear programming on the transmittance, has the defects of color distortion and dark brightness, so a gray world algorithm is proposed to balance the dark color difference, improve the brightness of the image, and improve the visual effect. In this embodiment, the step that the color of the original underwater image is corrected by using a gray world algorithm to obtain a second sharp image includes specifically includes the following steps: Al, a mean Gray of three channels R, (3, and B of the original underwater image is calculated: R+6-4 B Gra) = Where omy represents the mean of the three channels R, G, and B of the original 5 underwater image, and R-, J, and R-represent means of the R, G, and B channels, respectively.
A2, gain coefficients of the three channels R, G, and B of the original underwater image are calculated: Gruv Where k" kg, and ko represent the gain coefficients of the R, G, and B channels, respectively; A3, adaptive incremental adjustment is performed on the three channels R, G, and B of the original underwater image through color temperature correction: (7(R)= R*(o, C(0)=0* to, Where (7(R), C(0), and (7(H) respectively represent corresponding values after the three channels R, G, B of the original underwater image are multiplied by their respective gains; a) , ) , and 03 all represent gains; In this embodiment, the color temperature correction is to multiply the three channels of the image by their respective gains so as to adjust the ratio of three colors R, G, and B. The color temperature correction has to handle overflow, so as to prevent the output of R, G, and B from exceeding the range of 0 to 255, where the values of o>i, 6)2 and 6)3 are solved through the theory of maximum image information entropy according to the change of the image into a gray image, that is: max(H (R * co, + * ou, + B* co,)) (a),,co,,co,) Where H(x) represents an image entropy (one-dimensional entropy); and max(H (R * 01 + G * to2 + B * (03)) ((pi, represents the values of 01,0)2, 03 calculated when the image entropy is maximum. The one-dimensional entropy of the image represents the amount of information contained in the aggregate feature of gray distribution in the image. The one-dimensional gray entropy of the gray image is: ii = Where is the probability that a certain gray level appears in the image, which can be obtained from a gray level histogram.
A4, according to the VonKries diagonal model, for each pixel C in the image, its R, 10 G, and B components are adjusted: C(1)= C(R)* k, C = C(G) * kg (13)-0(13)* kb Where 0(R), C((2), and 0(B?) respectively represent the values obtained after C(R) , 0(G) , and (8) are multiplied by their respective gain coefficients, and k" k<, and k, respectively represent the gain coefficients of the R, G and B channels In a specific embodiment of the aforementioned underwater image enhancement method, further, the step that the first sharp image and the second sharp image are fused by using an image fusion algorithm based on wavelet transform to obtain a restored underwater image includes that: Three-layer wavelet decomposition is perfon led on the first sharp image and the second sharp image by using Haar wavelet basis to obtain low-frequency components (also referred to as low-frequency signals) and high-frequency components (also referred to as high-frequency signals) of different scales, wherein the low-frequency components (low-frequency signals) represent regions of the images where the brightness or gray value changes slowly (the amount of change is less than a preset change threshold), i.e., large flat regions in the images, which describe math parts of the images. The high-frequency components correspond to the parts where the brightness or gray value of the images change drastically (the amount of change is greater than or equal to the preset change threshold), that is, the edge (contour) or noise and details of the images The low-frequency components are processed by average operator fusion and multiplied by a gain coefficient (for example, 1.5 times) to obtain low-band wavelet 5 coefficients; The high-frequency components are processed by maximum regional energy fusion to obtain high-band wavelet coefficients; and An image is reconstructed by inverse wavelet transform according to the obtained low-band wavelet coefficients and high-band wavelet coefficients to complete the 10 fusion of the images and obtain the restored underwater image.
In this embodiment, the first sharp image and the second sharp image are fused by using an image fusion algorithm based on wavelet transform to obtain a restored underwater image with high quality and better visual effect, as shown in Fig. 2. Embodiment 2 The present invention also provides a specific embodiment of an underwater image enhancement device. The underwater image enhancement device provided by the present invention corresponds to the specific embodiment of the aforementioned underwater image enhancement method, and the underwater image enhancement device can perform the process steps in the specific embodiment of the method to achieve the purpose of the present invention, so the explanation in the specific embodiment of the underwater image enhancement method is also applicable to the specific embodiment of the underwater image enhancement device provided by the present invention, and will not be repeated in the following specific embodiment of the present invention As shown in Fig. 3, an embodiment of the present invention also provides an underwater image enhancement device, including: an establishment module 11, configured to acquire an original underwater image, and establish an underwater optical imaging model according to the characteristics of underwater imaging; a processing module 12, configured to sharpen the original underwater image by using the characteristic that the underwater optical imaging model is similar to an atmospheric model, and using a dark channel prior defogging algorithm that performs linear programming on the transmittance, to obtain a first sharp image, wherein the sharpening includes contrast enhancement; a correction module 13, configured to correct the color of the original underwater image by using a gray world algorithm to obtain a second sharp image; and a fusion module 14, configured to fuse the first sharp image and the second sharp image by using an image fusion algorithm based on wavelet transform to obtain a 10 restored underwater image.
According to the underwater image enhancement device described in the embodiment of the present invention, an original underwater image is acquired, and an underwater optical imaging model is established according to the characteristics of underwater imaging; the original underwater image is sharpened by using the characteristic that the underwater optical imaging model is similar to an atmospheric model, and using a dark channel prior defogging algorithm that performs linear programming on the transmittance, to obtain a first sharp image; the color of the original underwater image is corrected by using a gray world algorithm to obtain a second sharp image; and for the first dark sharp image corrected by the dark channel prior defogging algorithm and the second bright sharp image corrected by the gray world algorithm, the first sharp image and the second sharp image are fused by using an image fusion algorithm based on wavelet transform to obtain a restored underwater image with better contrast, color and brightness. In this way, the underwater image enhancement method based on multi-algorithm comprehensive optimization improves the visual effect of the underwater image, saves useful information of the image and reduces noise from the two aspects of contrast enhancement and color correction, so as to effectively improve the detail definition and color fidelity of the underwater image with low quality and low illumination, restore the real underwater scene, and then improve the accuracy and efficiency of underwater image restoration.
Described above are merely preferred embodiments of the present invention. It should be pointed out that many improvements and modifications may also be made for those of ordinary skill in the art without departing from the principle of the present invention, and these improvements and modifications shall fall into the protection scope of the present invention.
Claims (9)
- CLAIMS1. An underwater image enhancement method, comprising: acquiring an original underwater image, and establishing an underwater optical imaging model according to the characteristics of underwater imaging; sharpening the original underwater image by using the characteristic that the underwater optical imaging model is similar to an atmospheric model, and using a dark channel prior defogging algorithm that performs linear programming on the transmittance, to obtain a first sharp image, wherein the sharpening comprises contrast enhancement; correcting the color of the original underwater image by using a gray world algorithm to obtain a second sharp image; and fusing the first sharp image and the second sharp image by using an image fusion algorithm based on wavelet transform to obtain a restored underwater image.
- 2. The underwater image enhancement method according to claim 1, wherein the underwater optical imaging model is expressed as: /(x) = ../(x)* f(x) + A (1 -1(x)) where [(x) represents the original underwater image, .1(x) represents the first 20 sharp image, /(x) represents the transmittance of scene light, and A represents underwater ambient light.
- 3. The underwater image enhancement method according to claim 1, wherein the sharpening formula used by the dark channel prior defogging algorithm is: 25./(x) = /(x)-A +.1 lax(t(x),t0) where to represents a transmittance threshold.
- 4. The underwater image enhancement method according to claim 3. wherein 1(x) is newly linearly programmed to i(x) = 1(x)* 0.9 + 0.1 where represents the newly programmed transmittance.
- 5. The underwater image enhancement method according to claim 1, wherein sharpening the original underwater image by using the characteristic that the underwater optical imaging model is similar to an atmospheric model, and using a dark channel prior defogging algorithm that performs linear programming on the transmittance, to obtain a first sharp image comprises: down-sampling the original underwater image; calculating a dark channel and transmittance image by using the reduced image obtained by the down-sampling as input; and reconstructing the dark channel and transmittance image by means of bilinear interpolation to obtain the first sharp image with original image size. Is
- 6. The underwater image enhancement method according to claim 1, wherein correcting the color of the original underwater image by using a gray world algorithm to obtain a second sharp image comprises: multiplying three channels R, G, and B of the original underwater image by their 20 respective gains through color temperature correction to adjust the ratio of three colors R, G, and B to obtain: C(R)= 1?* a), C (G)= G* C (B)= B * co; where c(R) and co) respectively represent corresponding values after the three channels R, G, and B of the original underwater image are multiplied by their respective gains; "i, "2, and c), represent gains, and the values of, "2, and W3 are solved through the theory of maximum image information entropy according to the change of the image into a gray image: max(H(R * O+ G co, + B 3)) -> ("1."2' c° 3) , where 11(.) represents an image entropy; and adjusting, for each pixel C in the image, its R, G, and B components according to the obtained C(R) C(G), and c(B) : e(R')= c(R).k, C(C)=C(G)*kg C(13?)= *kb where c(k), (_7(0) , and c(i1) respectively represent the values obtained after c(R), C(G), and c(B) are multiplied by their respective gain coefficients, and k" 10 k, and kb respectively represent the gain coefficients of the R, G and B channels
- 7. The underwater image enhancement method according to claim 5, wherein k" , and kb are respectively expressed as: Gray where represents the mean of the three channels R, G, and B of the original underwater image, and R, G, and B represent means of the R, G, and B channels, respectively.
- 8. The underwater image enhancement method according to claim 1, wherein fusing the first sharp image and the second sharp image by using an image fusion algorithm based on wavelet transform to obtain a restored underwater image comprises: performing three-layer wavelet decomposition on the first sharp image and the second sharp image by using Haar wavelet basis to obtain low-frequency components and high-frequency components of different scales; processing the low-frequency components by average operator fusion and multiplying by a gain coefficient to obtain low-band wavelet coefficients; processing the high-frequency components by maximum regional energy fusion to obtain high-band wavelet coefficients; and reconstructing an image by inverse wavelet transform according to the obtained low-band wavelet coefficients and high-band wavelet coefficients to complete the fusion of the images and obtain the restored underwater image.
- 9. An underwater image enhancement device, comprising: an establishment module, configured to acquire an original underwater image, and establish an underwater optical imaging model according to the characteristics of underwater imaging; a processing module, configured to sharpen the original underwater image by using the characteristic that the underwater optical imaging model is similar to an atmospheric model, and using a dark channel prior defogging algorithm that perfouns linear programming on the transmittance, to obtain a first sharp image, wherein the sharpening comprises contrast enhancement; a correction module, configured to correct the color of the original underwater 20 image by using a gray world algorithm to obtain a second sharp image; and a fusion module, configured to fuse the first sharp image and the second sharp image by using an image fusion algorithm based on wavelet transform to obtain a restored underwater image.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910396668.5A CN110148095B (en) | 2019-05-14 | 2019-05-14 | Underwater image enhancement method and enhancement device |
Publications (3)
Publication Number | Publication Date |
---|---|
GB202006954D0 GB202006954D0 (en) | 2020-06-24 |
GB2585754A true GB2585754A (en) | 2021-01-20 |
GB2585754B GB2585754B (en) | 2022-06-01 |
Family
ID=67594214
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
GB2006954.8A Active GB2585754B (en) | 2019-05-14 | 2020-05-12 | Underwater image enhancement method and enhancement device |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN110148095B (en) |
GB (1) | GB2585754B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117036207A (en) * | 2023-10-10 | 2023-11-10 | 慧医谷中医药科技(天津)股份有限公司 | Method for enhancing infrared image in three-dimensional acquisition box |
Families Citing this family (30)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111080537B (en) * | 2019-11-25 | 2023-09-12 | 厦门大学 | Intelligent control method, medium, equipment and system for underwater robot |
CN111476744B (en) * | 2020-05-06 | 2023-05-16 | 北京理工大学 | Underwater image enhancement method based on classification and atmospheric imaging model |
CN112006651A (en) * | 2020-09-10 | 2020-12-01 | 孙礼华 | Cataract surgery auxiliary diagnosis system and method thereof |
CN112561804A (en) * | 2020-10-09 | 2021-03-26 | 天津大学 | Low-illumination underwater image enhancement method based on multi-scale detail enhancement |
CN112330548A (en) * | 2020-10-14 | 2021-02-05 | 北京理工大学珠海学院 | Underwater image restoration method, device, equipment and medium based on multi-scale fusion |
CN112907470A (en) * | 2021-02-05 | 2021-06-04 | 北京理工大学 | Underwater image recovery method based on Lab color gamut transformation, classification and white balance |
CN113256563A (en) * | 2021-04-25 | 2021-08-13 | 广东工业大学 | Method and system for detecting surface defects of fine product tank based on space attention mechanism |
CN113284060B (en) * | 2021-05-17 | 2024-04-05 | 大连海事大学 | Underwater image enhancement method based on wavelength attenuation identification |
CN113240606B (en) * | 2021-05-24 | 2024-06-21 | 博奥生物集团有限公司 | Traditional Chinese medicine inspection image color correction method and system |
CN113487501A (en) * | 2021-06-29 | 2021-10-08 | 嵊州市浙江工业大学创新研究院 | Novel underwater image restoration method based on dark channel prior and BM3D algorithm |
CN113538309A (en) * | 2021-07-09 | 2021-10-22 | 北京理工大学 | Multi-scale fusion restoration method and device for underwater scattering image |
CN113538543B (en) * | 2021-07-15 | 2024-04-30 | 大连海事大学 | Underwater image restoration method based on texture distribution and light absorption |
CN113658067B (en) * | 2021-08-11 | 2022-08-12 | 沭阳天勤工具有限公司 | Water body image enhancement method and system in air tightness detection based on artificial intelligence |
CN113657497B (en) * | 2021-08-17 | 2024-03-08 | 上海亨临光电科技有限公司 | Method for realizing low signal-to-noise ratio image enhancement processing and human body image segmentation fusion |
CN113962890B (en) * | 2021-10-22 | 2024-05-10 | 大连海事大学 | Affine transformation-based underwater color cast image sharpening method |
CN114202491B (en) * | 2021-12-08 | 2022-09-09 | 深圳市研润科技有限公司 | Method and system for enhancing optical image |
CN114627001B (en) * | 2022-02-08 | 2024-09-06 | 南京理工大学 | Underwater robot image restoration method based on dark channel prior algorithm |
CN114612354B (en) * | 2022-03-01 | 2024-08-27 | 南京理工大学 | HSV color space-based multi-scale fusion underwater image enhancement method |
CN114897710A (en) * | 2022-04-06 | 2022-08-12 | 四川大学 | Video enhancement method under low-illumination environment |
CN114821021A (en) * | 2022-04-28 | 2022-07-29 | 昆明理工大学 | Underwater image enhancement method combining multichannel equalization and multi-scale fusion |
CN115034981B (en) * | 2022-05-27 | 2024-08-06 | 大连海事大学 | Deep learning underwater image enhancement method based on embedded multi-feature fusion |
CN115861450A (en) * | 2022-10-28 | 2023-03-28 | 河南科技学院 | Piecewise linear stretching underwater image color correction method |
CN116071520B (en) * | 2023-03-31 | 2023-07-14 | 湖南省水务规划设计院有限公司 | Digital twin water affair simulation test method |
CN116309232B (en) * | 2023-05-15 | 2023-08-18 | 中国海洋大学 | Underwater image enhancement method combining physical priori with deep learning |
CN116863320B (en) * | 2023-06-06 | 2024-03-29 | 山东大学 | Underwater image enhancement method and system based on physical model |
CN116631038B (en) * | 2023-06-06 | 2024-05-28 | 湖南三湘银行股份有限公司 | Method and system for verifying identity of bank user based on image |
CN116523801B (en) * | 2023-07-03 | 2023-08-25 | 贵州医科大学附属医院 | Intelligent monitoring method for nursing premature infants |
CN116934833B (en) * | 2023-07-18 | 2024-08-13 | 广州大学 | Binocular vision-based underwater structure disease detection method, equipment and medium |
CN116823674B (en) * | 2023-08-24 | 2024-03-12 | 湖南省水务规划设计院有限公司 | Cross-modal fusion underwater image enhancement method |
CN117474818B (en) * | 2023-12-27 | 2024-03-15 | 北京科技大学 | Underwater image enhancement method and device based on non-parameter Bayesian depth of field estimation |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942758B (en) * | 2014-04-04 | 2017-02-15 | 中国人民解放军国防科学技术大学 | Dark channel prior image dehazing method based on multiscale fusion |
CN106600547A (en) * | 2016-11-17 | 2017-04-26 | 天津大学 | Underwater image restoration method |
CN107610058A (en) * | 2017-08-28 | 2018-01-19 | 浙江工业大学 | High-definition picture defogging method based on down-sampling |
CN109389579A (en) * | 2018-08-29 | 2019-02-26 | 江苏理工学院 | A kind of combination is entropy constrained anti-light according to interference method with the textile acetes chinensis of standard deviation weighting |
-
2019
- 2019-05-14 CN CN201910396668.5A patent/CN110148095B/en active Active
-
2020
- 2020-05-12 GB GB2006954.8A patent/GB2585754B/en active Active
Non-Patent Citations (2)
Title |
---|
Computer Vision and Understanding, vol. 165, 2017, Li Yu et al., "Haze visibility enhancement: A Survey and quantitative benchmarking", pages 1-16. * |
Rajesh, S.D. et al, June 2019, "Image Cleaning and Enhancement Technique for Underwater Mining", OCEANS 2019-Marseille (pp. 1-6), IEEE. * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117036207A (en) * | 2023-10-10 | 2023-11-10 | 慧医谷中医药科技(天津)股份有限公司 | Method for enhancing infrared image in three-dimensional acquisition box |
CN117036207B (en) * | 2023-10-10 | 2024-01-19 | 慧医谷中医药科技(天津)股份有限公司 | Method for enhancing infrared image in three-dimensional acquisition box |
Also Published As
Publication number | Publication date |
---|---|
CN110148095B (en) | 2021-05-07 |
CN110148095A (en) | 2019-08-20 |
GB2585754B (en) | 2022-06-01 |
GB202006954D0 (en) | 2020-06-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
GB2585754A (en) | Underwater image enhancement method and enhancement device | |
CN105761227B (en) | Underwater picture Enhancement Method based on dark channel prior and white balance | |
Shi et al. | Let you see in sand dust weather: A method based on halo-reduced dark channel prior dehazing for sand-dust image enhancement | |
CN111292258B (en) | Image defogging method based on dark channel prior and bright channel prior | |
CN105631829B (en) | Night haze image defogging method based on dark channel prior and color correction | |
CN112419162B (en) | Image defogging method, device, electronic equipment and readable storage medium | |
CN102789635B (en) | Image enhancement method and image enhancement device | |
GB2581907A (en) | Image enhancement method and system | |
CN108765342A (en) | A kind of underwater image restoration method based on improvement dark | |
CN105303532A (en) | Wavelet domain Retinex image defogging method | |
CN106780368A (en) | A kind of underwater picture Enhancement Method based on foreground model | |
CN109658343B (en) | Underwater image enhancement method combining color conversion and adaptive exposure | |
CN107527325B (en) | Monocular underwater vision enhancement method based on dark channel priority | |
CN107862672B (en) | Image defogging method and device | |
Yu et al. | An improved multi-scale retinex fog and haze image enhancement method | |
CN109272475B (en) | Method for rapidly and effectively repairing and strengthening underwater image color | |
CN104899836A (en) | Foggy image enhancing device and method based on near infrared multispectral imaging | |
Mathur et al. | Enhancement of underwater images using white balancing and Rayleigh-stretching | |
CN108093175A (en) | A kind of adaptive defogging method of real-time high-definition video and device | |
CN110782400B (en) | Self-adaptive illumination uniformity realization method and device | |
Asari et al. | Nonlinear enhancement of extremely high contrast images for visibility improvement | |
Raigonda et al. | Haze Removal Of Underwater Images Using Fusion Technique | |
CN107317968B (en) | Image defogging method and device, computer storage medium and mobile terminal | |
CN115170404A (en) | Night image defogging method and device based on brightness compensation and layer separation | |
Nan et al. | Image defogging algorithm based on Fisher criterion function and dark channel prior |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
732E | Amendments to the register in respect of changes of name or changes affecting rights (sect. 32/1977) |
Free format text: REGISTERED BETWEEN 20231116 AND 20231122 |